New Measures of Integrated Information for Time-Series Data
Author Information
Author(s): Barrett Adam B., Seth Anil K.
Primary Institution: Sackler Centre for Consciousness Science and School of Informatics, University of Sussex, Brighton, United Kingdom
Hypothesis
Can new measures of integrated information be effectively applied to time-series data in biological systems?
Conclusion
The new measures of integrated information provide practical tools for analyzing complex biological systems and their implications for consciousness.
Supporting Evidence
- The new measures can be applied to both discrete and continuous systems.
- ΦE and ΦAR are designed to overcome limitations of previous measures that were only applicable to discrete Markovian systems.
- Simulations showed that the new measures align with intuitions about information integration in neural systems.
Takeaway
This study introduces two new ways to measure how different parts of a system work together to create information, which can help us understand how the brain works.
Methodology
The study developed two new measures, ΦE and ΦAR, to quantify integrated information in time-series data, using simulations to demonstrate their applicability.
Limitations
The measures may exhibit instabilities due to normalization processes and are primarily applicable to stationary systems.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website